Visible Infrared Image Registration via 3D Coordinate Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods face challenges in accurately registering visible and infrared images of structures due to misregistration issues, where the positions of points other than reference points do not coincide, leading to incorrect associations and increased false detections in defect diagnosis.
Innovation Solution
An image processing device and method that acquires visible and infrared images of the same object from different positions, estimates the positions of non-reference points based on reference points, and associates the image values to reduce misregistration by projecting non-reference points onto the imaging surfaces, allowing for accurate superimposition and improved defect discrimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visible image and infrared image are superimposed using conventional geometric correction, then reference points can be aligned, but non-reference points remain misregistered
Solution Approach 1:
The patent introduces a 3D coordinate system as an intermediary reference frame to transform image coordinates from both visible and infrared cameras. By projecting image points onto the 3D object surface and then mapping to a standardized 2D coordinate system, the patent achieves accurate correspondence between non-reference points in both images, resolving the misregistration issue that plagues conventional methods.
2Area of stationary object
If images are captured from different positions with different angles of view, then comprehensive coverage is achieved, but misregistration between images occurs
Solution Approach 1:
The patent transitions from 2D image coordinates to 3D object surface coordinates as an intermediate representation. By introducing the third dimension (depth/distance from camera to object surface), the patent can accurately map points from different viewing angles onto the same 3D surface, enabling precise 2D projection and alignment while maintaining comprehensive imaging coverage.
3Device complexity
If misregistration remains between visible and infrared images, then independent analysis is simple, but defect discrimination performance deteriorates
Solution Approach 1:
The patent segments the image registration process into distinct stages: (1) capturing images from multiple cameras, (2) transforming coordinates to a unified 3D object surface system, (3) projecting to standardized 2D coordinates, and (4) performing defect analysis. This segmentation allows for accurate multi-image registration while maintaining clear processing logic and enabling effective defect discrimination through coordinated analysis.
Data Source
AI summary
There are provided an image processing device, an image processing method, and an image processing program that can reduce misregistration in associating a plurality of images. In an image processing device according to an aspect of the present invention, a processor acquires a first image and a second image in which the same object is imaged from different positions and which are two-dimensional images captured in different wavelength ranges, acquires information indicating positions of reference points on a surface of the object, and associates values of the first image with values of the second image on the basis of the acquired first image, the acquired second image, and the acquired information. The values of the first image and the values of the second image correspond to non-reference points that are points other than the reference points on the surface.